splitDataToTrainTestStratified: Function to do stratified sampling for model data

In ivanliu1989/RQuant: A collection of generic functions

Description

Function to stratified split a dataset and return training, test and eval sets as data.frame or data.table

Usage

splitDataToTrainTestStratified(x, strataCols = NULL, trainProportion,
  evalProportion = NULL, levels = 50, seed = 999, DT = T)

Arguments

x	A data.frame or data.table object containing all data to be split
strataCols	Names of columns or features that will be used to do stratification
trainProportion	Proportion of training set
evalProportion	Proportion of validation set
levels	Threshold of determine a feature to be categorical and qualified for stratification
seed	Random seeds for reproducibility
DT	If the function returns a list of data.tables or data.frames

Value

A list of data.table or data.frame

Examples

data(mtcars)
res <- splitDataToTrainTestStratified(mtcars, names(mtcars), 0.7, 0.3, 15, 999, T)
res$trainData
res$testData
res$evalData

ivanliu1989/RQuant documentation built on Sept. 13, 2019, 11:53 a.m.